CVE-2026-89985

A flaw was found in the Linux kernel's memory cgroup (memcg) component. This vulnerability arises when memory pages (folios) are moved between different memory nodes, causing an inconsistency in how their associated object cgroups (objcg) are managed. This can lead to memory management operations acquiring incorrect locks, potentially resulting in system instability or a denial of service (DoS).

A race in memcg folio migration and objcg reparenting can break the invariant that an LRU folio uses the objcg for its own NUMA node. After migration across nodes, reparenting of the objcg and reparenting of the folio LRU list can happen in different lock sections, so an LRU operation may resolve the wrong lruvec and take the wrong lru_lock for the list that actually contains the folio. This can strand folios on the LRU list of a dying memcg and creates a kernel memory management corruption candidate with at least local DoS risk. For the CVSS the PR:L is used for the paranoid score because delegated cgroup control, container like environments, or service accounts may allow a local actor to influence memcg lifetime, memory pressure, and migration paths without full host root. The issue is not network reachable. Impact is at least a local denial of service and in worst case may allow confidentiality or integrity impact if the stale LRU or dead memcg state becomes exploitable as a UAF style primitive.

The CVSS score(s) listed for this vulnerability do not reflect the associated product's status, and are included for informational purposes to better understand the severity of this vulnerability.

Mitigation for this issue is either not available or the currently available options don't meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base or stability.